| Bit-tactical: A software/hardware approach to exploiting value and bit sparsity in neural networks A Delmas Lascorz, P Judd, DM Stuart, Z Poulos, M Mahmoud, S Sharify, ... Proceedings of the Twenty-Fourth International Conference on Architectural …, 2019 | 162 | 2019 |
| Laconic deep learning inference acceleration S Sharify, AD Lascorz, M Mahmoud, M Nikolic, K Siu, DM Stuart, Z Poulos, ... Proceedings of the 46th International Symposium on Computer Architecture …, 2019 | 154 | 2019 |
| Memory requirements for convolutional neural network hardware accelerators K Siu, DM Stuart, M Mahmoud, A Moshovos 2018 IEEE International Symposium on Workload Characterization (IISWC), 111-121, 2018 | 109 | 2018 |
| Tensordash: Exploiting sparsity to accelerate deep neural network training M Mahmoud, I Edo, AH Zadeh, OM Awad, G Pekhimenko, J Albericio, ... 2020 53rd Annual IEEE/ACM International Symposium on Microarchitecture …, 2020 | 104 | 2020 |
| Diffy: A Déjà vu-free differential deep neural network accelerator M Mahmoud, K Siu, A Moshovos 2018 51st Annual IEEE/ACM International Symposium on Microarchitecture …, 2018 | 84 | 2018 |
| Shapeshifter: Enabling fine-grain data width adaptation in deep learning AD Lascorz, S Sharify, I Edo, DM Stuart, OM Awad, P Judd, M Mahmoud, ... Proceedings of the 52nd Annual IEEE/ACM International Symposium on …, 2019 | 51 | 2019 |
| Mokey: Enabling narrow fixed-point inference for out-of-the-box floating-point transformer models AH Zadeh, M Mahmoud, A Abdelhadi, A Moshovos Proceedings of the 49th Annual International Symposium on Computer …, 2022 | 43 | 2022 |
| Characterizing sources of ineffectual computations in deep learning networks M Nikolić, M Mahmoud, A Moshovos, Y Zhao, R Mullins 2019 IEEE International Symposium on Performance Analysis of Systems and …, 2019 | 32 | 2019 |
| Bit-tactical: Exploiting ineffectual computations in convolutional neural networks: Which, why, and how A Delmas, P Judd, DM Stuart, Z Poulos, M Mahmoud, S Sharify, M Nikolic, ... arXiv preprint arXiv:1803.03688, 2018 | 28 | 2018 |
| FPRaker: A processing element for accelerating neural network training OM Awad, M Mahmoud, I Edo, AH Zadeh, C Bannon, A Jayarajan, ... MICRO-54: 54th Annual IEEE/ACM International Symposium on Microarchitecture …, 2021 | 23 | 2021 |
| IDEAL: Image denoising accelerator M Mahmoud, B Zheng, AD Lascorz, F Heide, J Assouline, P Boucher, ... Proceedings of the 50th Annual IEEE/ACM International Symposium on …, 2017 | 19 | 2017 |
| Boveda: Building an on-chip deep learning memory hierarchy brick by brick I Edo Vivancos, S Sharify, D Ly-Ma, A Abdelhadi, C Bannon, M Nikolic, ... Proceedings of Machine Learning and Systems 3, 1-20, 2021 | 12 | 2021 |
| Laconic deep learning computing S Sharify, M Mahmoud, AD Lascorz, M Nikolic, A Moshovos arXiv preprint arXiv:1805.04513, 2018 | 4 | 2018 |
| Accelerating Image-Sensor-Based Deep Learning Applications M Mahmoud, DM Stuart, Z Poulos, AD Lascorz, P Judd, S Sharify, ... IEEE Micro 39 (5), 26-35, 2019 | 2 | 2019 |
| Identifying and Exploiting Ineffectual Computations to Enable Hardware Acceleration of Deep Learning A Moshovos, J Albericio, P Judd, A Delmas, S Sharify, M Mahmoud, ... 2018 16th IEEE International New Circuits and Systems Conference (NEWCAS …, 2018 | 2 | 2018 |
| Memory controller design under cloud workloads M Mahmoud, A Moshovos 2016 IEEE International Symposium on Workload Characterization (IISWC), 1-11, 2016 | 2 | 2016 |
| Hybrid limited-pointer linked-list cache directory and cache coherence protocol M Mahmoud, A Wassal JEC-ECC 2013, 77 - 82, 2013 | 2 | 2013 |
| Building an on-chip deep learning memory hierarchy brick by brick: late breaking results IE Vivancos, S Sharify, M Nikolic, C Bannon, M Mahmoud, AD Lascorz, ... Proceedings of the 57th ACM/EDAC/IEEE Design Automation Conference, 1-2, 2020 | 1 | 2020 |
| Tensordash: Exploiting sparsity to accelerate deep neural network training and inference M Mahmoud, IE Vivancos, O Awad, AH Zadeh, G Pekhimenko, J Albericio, ... Arxiv preprint cs. AR, 0 | 1 | |
| Method and device with convolution neural network processing M MAHMOUD, A Moshovos US Patent 11,836,971, 2023 | | 2023 |
Andreas MoshovosProfessor, Electrical and Computer Engineering, University of TorontoVerified email at utoronto.ca
